Instance grid · atlas

Each row is an instance family, each column a size. One step right doubles the vCPUs, the memory and the price, so the cheapest family for your workload is the lightest row.

Cheapest for its sizePriciest

General purpose 4 GiB memory per vCPU
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
m8g Graviton4	$24	$48	$96	$193	$385	$771	$1.2k	$1.5k	$2.3k		$4.6k
m7g Graviton3	$22	$44	$88	$175	$350	$701	$1.1k	$1.4k				$1.4k
m7i Xeon Sapphire Rapids		$54	$108	$216	$433	$865	$1.3k	$1.7k	$2.6k		$5.2k
m7i-flex Xeon Sapphire Rapids		$51	$103	$206	$411	$822	$1.2k	$1.6k
m6g Graviton2	$20	$39	$79	$157	$315	$630	$945	$1.3k				$1.3k
m6gd Graviton2	$23	$46	$93	$185	$371	$742	$1.1k	$1.5k				$1.5k
m6i Xeon Ice Lake		$52	$103	$206	$412	$824	$1.2k	$1.6k	$2.5k	$3.3k		$3.3k
m6id Xeon Ice Lake		$61	$122	$243	$487	$974	$1.5k	$1.9k	$2.9k	$3.9k		$3.9k
m6idn Xeon Ice Lake		$80	$160	$320	$641	$1.3k	$1.9k	$2.6k	$3.8k	$5.1k		$5.1k
m6in Xeon Ice Lake		$71	$143	$285	$571	$1.1k	$1.7k	$2.3k	$3.4k	$4.6k		$4.6k
m6a EPYC Milan		$46	$93	$185	$371	$742	$1.1k	$1.5k	$2.2k	$3.0k	$4.5k	$4.5k
m5 Xeon Skylake Newest		$49	$99	$197	$393	$788	$1.2k	$1.6k	$2.4k			$2.4k
m5d Xeon Skylake		$58	$116	$232	$464	$927	$1.4k	$1.9k	$2.8k			$2.8k
m5a EPYC Naples		$44	$88	$177	$353	$707	$1.1k	$1.4k	$2.1k
m5ad EPYC Naples		$53	$106	$212	$423	$846	$1.3k	$1.7k	$2.5k
m4 Xeon Haswell		$50	$100	$199	$399			$1.6k
Compute optimized 2 GiB memory per vCPU
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
c8g Graviton4	$21	$42	$84	$169	$337	$674	$1.0k	$1.3k	$2.0k		$4.0k
c7g Graviton3	$19	$38	$77	$153	$307	$613	$920	$1.2k				$1.2k
c7gd Graviton3	$23	$46	$92	$184	$369	$737	$1.1k	$1.5k				$1.5k
c7i Xeon Sapphire Rapids		$47	$95	$189	$379	$757	$1.1k	$1.5k	$2.3k		$4.5k
c7i-flex Xeon Sapphire Rapids		$45	$90	$180	$360	$719	$1.1k	$1.4k
c6g Graviton2	$17	$34	$68	$137	$274	$548	$821	$1.1k				$1.1k
c6gd Graviton2	$19	$39	$78	$156	$312	$624	$936	$1.2k				$1.2k
c6gn Graviton2	$22	$43	$87	$173	$347	$694	$1.0k	$1.4k				$1.4k
c6i Xeon Ice Lake		$45	$90	$180	$361	$721	$1.1k	$1.4k	$2.2k	$2.9k		$2.9k
c6id Xeon Ice Lake		$51	$102	$205	$409	$819	$1.2k	$1.6k	$2.5k	$3.3k		$3.3k
c6in Xeon Ice Lake		$57	$114	$228	$456	$912	$1.4k	$1.8k	$2.7k	$3.6k		$3.6k
c6a EPYC Milan		$41	$81	$162	$325	$649	$974	$1.3k	$1.9k	$2.6k	$3.9k	$3.9k
c5 Xeon Skylake Newest		$43	$85	$171	$342		$1.0k		$2.1k			$2.1k
c5d Xeon Skylake		$49	$98	$195	$390		$1.2k		$2.3k			$2.3k
c5n Xeon Skylake		$54	$109	$217	$434							$2.0k
c5a EPYC Rome		$39	$77	$155	$309	$618	$927	$1.2k	$1.9k
c4 Xeon Haswell		$54	$107	$215	$429	$858
Memory optimized 8 GiB or more memory per vCPU
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
r8g Graviton4	$31	$62	$125	$250	$500	$999	$1.5k	$2.0k	$3.0k		$6.0k
r7g Graviton3	$28	$57	$114	$227	$454	$908	$1.4k	$1.8k				$1.8k
r7gd Graviton3	$34	$69	$137	$275	$549	$1.1k	$1.6k	$2.2k				$2.2k
r7i Xeon Sapphire Rapids		$70	$140	$281	$561	$1.1k	$1.7k	$2.2k	$3.4k		$6.7k
r6g Graviton2	$25	$51	$102	$203	$406	$812	$1.2k	$1.6k				$1.6k
r6gd Graviton2	$29	$58	$116	$233	$465	$930	$1.4k	$1.9k				$1.9k
r6i Xeon Ice Lake		$67	$134	$267	$534	$1.1k	$1.6k	$2.1k	$3.2k	$4.3k		$4.3k
r6idn Xeon Ice Lake		$99	$198	$395	$791	$1.6k	$2.4k	$3.2k	$4.7k	$6.3k		$6.3k
r6in Xeon Ice Lake		$88	$176	$352	$703	$1.4k	$2.1k	$2.8k	$4.2k	$5.6k		$5.6k
r6a EPYC Milan		$60	$120	$240	$481	$962	$1.4k	$1.9k	$2.9k	$3.8k	$5.8k	$5.8k
r5 Xeon Skylake Newest		$64	$127	$254	$508	$1.0k	$1.5k	$2.0k	$3.0k			$3.0k
r5b Xeon Cascade Lake		$75	$150	$300	$601	$1.2k	$1.8k	$2.4k	$3.6k			$3.6k
r5d Xeon Skylake		$73	$145	$291	$581	$1.2k	$1.7k	$2.3k	$3.5k			$3.5k
r5n Xeon Cascade Lake		$75	$150	$301	$601	$1.2k	$1.8k	$2.4k	$3.6k			$3.6k
r5a EPYC Naples		$57	$114	$228	$456	$913	$1.4k	$1.8k	$2.7k
r5ad EPYC Naples		$66	$132	$265	$530	$1.1k	$1.6k	$2.1k	$3.2k
r4 Xeon Broadwell		$67	$135	$270	$540	$1.1k		$2.2k
x2idn Xeon Ice Lake								$3.3k	$4.9k	$6.6k		$6.6k
x2iedn Xeon Ice Lake			$412	$824	$1.6k	$3.3k		$6.6k	$9.9k	$13k		$13k
x1 Xeon Haswell								$3.3k		$6.6k
x1e Xeon Haswell			$412	$825	$1.6k	$3.3k		$6.6k		$13k
Storage optimized Fast local disks attached to the host
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
i4g Graviton2		$81	$162	$324	$649	$1.3k		$2.6k
im4gn Graviton2		$94	$189	$377	$755	$1.5k		$3.0k
is4gen Graviton2	$79	$159	$318	$635	$1.3k	$2.5k
i4i Xeon Ice Lake		$90	$180	$360	$721	$1.4k	$2.2k	$2.9k	$4.3k	$5.8k		$5.8k
d3 Xeon Cascade Lake Newest			$253	$506	$1.0k	$2.0k
i3 Xeon Broadwell Newest		$86	$172	$343	$687	$1.4k		$2.7k
i3en Xeon Skylake		$125	$249	$499			$3.0k		$6.0k			$6.0k
Accelerated computing GPUs and AWS ML chips
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
g6 EPYC Milan · 24 GB per GPU			$425	$516	$698	$1.1k	$2.4k ×4	$1.8k	$3.5k ×4		$7.0k ×8
gr6 EPYC Milan · 24 GB per GPU					$812	$1.3k
g5 EPYC Rome · 24 GB per GPU			$514	$619	$829	$1.3k	$2.9k ×4	$2.1k	$4.2k ×4		$8.3k ×8
g4dn Xeon Family · 16 GB per GPU			$269	$384	$615	$1.1k	$2.0k ×4	$2.2k				$4.0k ×8
p4d Xeon Cascade Lake · 40 GB per GPU									$12k ×8
inf1 Xeon Cascade Lake			$117	$185					$2.4k ×16
Burstable Cheap baseline CPU that bursts on credits
	nano	micro	small	medium	large	xlarge	2xlarge
t4g Graviton2	$2.12	$4.23	$8.47	$17	$34	$68	$135
t3 Xeon Skylake Newest	$2.70	$5.33	$11	$21	$43	$85	$171
t3a EPYC Naples	$2.41	$4.82	$9.64	$19	$38	$77	$154
t2 Xeon Family	$2.92	$5.84	$12	$23	$47	$93	$186